CUDA transfer memory during kernel execution

Question

I know that CUDA kernels can be "overlapped" by putting them into separate streams, but I'm wondering if would it be possible to transfer memory during kernel executions. CUDA kernels are asynchronous afterall

score 2 · Accepted Answer · answered Apr 19 '12 at 22:23

2

You can run kernels, transfers from host to device and transfers from device to host concurrently.

http://developer.download.nvidia.com/CUDA/training/StreamsAndConcurrencyWebinar.pdf

answered Apr 19 '12 at 22:23

Roger Dahl

15,132
8
62
82

score 1 · Answer 2 · answered Apr 20 '12 at 16:07

1

Just for clarification, the above are valid only if your device supports it. You can check it running device query and checking the attribute concurrent copy and execution

answered Apr 20 '12 at 16:07

amanda

394
1
9

Thank you for the information, I knew it but it's always useful to repeat it – paulAl Apr 21 '12 at 11:07

CUDA transfer memory during kernel execution

2 Answers2